Skip to content

Latest commit

 

History

History
16 lines (12 loc) · 1023 Bytes

README.md

File metadata and controls

16 lines (12 loc) · 1023 Bytes
Raw

Optical Character Recognition

This repository contains implementation for Optical Character Recognition (OCR) using Python. The code is specifically designed to detect Iranian ID cards and credit cards and extract their numbers.

Usage

  • Open the file in Google Colab
  • Run all the sections sequentially
  • Pay attention that you should run the Data Collection section only once
  • Between the Load Model and Train section you can choose either to load the pre-trained model (existing model in this repository), or train your own network
  • The prediction section shows you the accuracy on the validation data
  • In the Prediction Function section you can give the address of directory and see the result and predicted labels of the trained model (for any desired input and image)
  • That's all !

This is the link to a complete document about this project. Hope you find these instructions useful ✌😊 https://docs.google.com/document/d/1_pKAtI2W-8YWLEnv54Uu_npd2rAv8KUY7dYxu65yBdI/edit?usp=sharing